www.nature.

com/scientificreports

OPEN Evaluation
of the microencapsulation
of orange essential oil
in biopolymers by using
a spray‑drying process
Maria Clara Santana Aguiar, Maria Fátima das Graças Fernandes da Silva,
João Batista Fernandes & Moacir Rossi Forim*

Essential oils are volatile compounds commonly used by several industries, easily degradable, which
restrains their applications. Therefore, we developed and validated a methodology for producing
microcapsules loaded with orange essential oil, using a spray-drying process. The experimental
design results showed that the combination between a low flow transfer rate (0.15 L h−1) of the
colloidal suspension, a higher drying air flow rate (536 L h−1), and an inlet air temperature of 150 °C
to the spray-dryer were the most important parameters for the atomization efficiency. The method
optimization resulted in microcapsules with powder recovery between 7.6 and 79.9% (w w−1), oil
content ranging from 8.9 to 90.4% (w w−1), encapsulation efficiency between 5.7 and 97.0% (w w−1),
and particle sizes with a high frequency of distribution less than 4 μm. In these experiments, gelatin
and lignin were evaluated as biopolymers of encapsulation. We also developed an analytical method
using headspace gas chromatography. The matrix effects could be addressed by using matrix-matched
calibration curves. The chromatographic analysis was linear and selective for d-limonene between
0.025 and 3.00 µg mL−1, with correlation coefficients higher than 0.99. The analytical method had
limits of detection and quantitation of 0.024 and 0.073 mg g−1 for gelatin and 0.039 and 0.119 mg g−1
for lignin, respectively.

Essential oils (EOs) are botanical secondary metabolites of low molecular ­mass1. These compounds have gained
considerable attention because of their applications in cosmetics, detergents, medicines, food production, and
pest ­protection2,3. Among EOs, orange essential oil stands out since it is an abundant coproduct from citrus
processing that has a characteristic flavor and fragrance and has antimicrobial, antifungal, and antioxidant
­activities4–6.
However, few EOs are water-soluble and they are easily degraded by heat, oxidation, and ­light7. To improve
their dispersion in an aqueous medium, biological action, and protection from environmental agents, method-
ologies that promote the encapsulation of EOs have been ­developed7–9. Among the methods for encapsulating
essential oils, spray-drying has received attention because of its low operational cost, compatibility with labile
materials, it can produce stable final products, and compatibility with continuous large-scale p ­ roduction10,11.
Because of the multiple variables that can influence the spray-drying process, the use of biopolymers as wall
material is a research area of increasing interest. In this group, gelatin and lignin can be ideal candidates. In
addition to their structural characteristics and ability to self-associate10,12, these products are biodegradable and
biocompatible ­materials12,13. Moreover, it has a low cost that is favorable from a commercial point of view. The
lignin is a waste product of the pulp and paper industry.
Some studies have been conducted to evaluate the encapsulation of orange essential oil using spray-drying
process. In these studies, a variety of polymers such as modified starch, maltodextrin, whey protein, cellulose
nanofibrils, and Arabic gum were reported as showing potential for application in atomization drying p ­ rocess14–16.

Department of Chemistry, Federal University of São Carlos, Rod. Washington Luiz, Km 235, São Carlos, SP 13565‑905,
Brazil. *email: mrforim@ufscar.br

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 1

Vol.:(0123456789)
 www.nature.com/scientificreports/

Powder recovery Encapsulation

(%) Oil ­contenta (%) ­efficiencya (%) Global desirability
Experiment Gelatin Lignin Gelatin Lignin Gelatin Lignin Gelatin Lignin
1 21.59 31.16 62.36 19.06 37.20 49.55 0.49 0.34
2 41.63 44.52 32.28 24.88 61.24 50.07 0.54 0.58
3 17.58 16.88 20.39 11.50 10.40 5.69 0.19 0.06
4 13.67 25.60 65.56 11.13 44.97 18.16 0.42 0.12
5 10.98 52.11 64.80 30.96 34.10 46.77 0.38 0.78
6 41.13 53.29 26.94 13.91 33.66 13.44 0.50 0.40
7 43.23 79.85 10.94 8.85 21.51 13.43 0.43 0.50
8 18.54 59.02 90.41 15.62 46.69 25.10 0.63 0.49
9 32.30 43.30 9.47 13.28 15.40 11.49 0.29 0.31
10 21.18 33.96 80.54 27.65 49.48 96.72 0.60 0.56
11 15.09 32.66 71.20 12.58 48.38 18.64 0.47 0.21
12 25.68 34.96 13.13 12.55 9.37 12.44 0.24 0.23
13 42.78 62.34 13.61 13.98 16.03 24.16 0.44 0.48
14 44.25 72.70 53.47 23.62 97.00 28.93 0.70 0.78
15 7.56 50.48 86.40 13.92 19.67 44.68 0.48 0.38
16 49.98 54.13 9.84 20.14 23.37 41.79 0.50 0.55

Table 1.  Influence of the used biopolymer (gelatin and lignin) on power recovery, oil content, and
encapsulation efficiency. Experiments 1 to 16 refer to the fractioned factorial design.  The values from 1 to 16
refer to the experiments proposed by the factorial design; % in w w−1; a values referent to d-limonene.

However, there is little information available on the encapsulation of orange essential oil using gelatin or lignin
as biopolymers.
Despite the promising data, there is a lack of adequate information in the literature that not only describes
validated methodologies for the encapsulation of essential oils but also describes reliable analytical methods that
meet the requirements of analytical applications, especially for the analysis of atomization efficiency. Quantita-
tive analysis in quality control of the formulated essential oil is important in determining the features of the
final product (reproducibility, stability, dispersion, release kinetics, etc.), as well as its relationship to the desired
biological activities.
Considering the need to develop methods for the encapsulation of EOs and quality control analyses, the main
aims of this present work were to propose a method using spray-drying for the encapsulation of orange essential
oil and its subsequent quality control analysis through quantification by Headspace Gas Chromatography with
Flame Ionization Detector (HS-GC-FID). Moreover, the developed analytical method was validated through
different parameters, and the essential oil was characterized by gas chromatography-mass spectrometry.

Results and discussion
Matrix characterization.  The volatile constituents of the orange EO, their retention time, and the relative
area corresponding to each compound are described in Supplementary Table S1. In this analysis, we verified the
presence of 27 different compounds and identified 24. We chose to highlight d-limonene (77.5%), β-myrcene
(11.1%), α-pinene (3.99%) and linalool (1.44%) because these four constituents represented 94% of the total
relative abundance area. These terpenes were also found to be the major constituents when orange essential
oil was extracted by a Clevenger ­apparatus4, supercritical ­fluid17, and microwave steam ­distillation18, and they
were related to the biological activity of orange EO. Since d-limonene was the major component identified, we
selected it as a quality control marker for the development of an analytical method to be used to determine the
atomization efficiency.

Development of orange essential oil encapsulation methodology.  One of the great difficulties
in working with essential oils is to prevent the loss of its volatile constituents. Moreover, for microencapsulated
products loaded with essential oil, we also face problems of the compatibility of the polymers with the chroma-
tographic analyses. In most nano- and microencapsulation protocols such as spray-drying, heating steps can
result in the partial loss of the essential oil, resulting in both economical losses and difficulties in interpreting the
results of the biological assays. In this context, the optimization and analysis of EO content are critical.
Thus, to optimize the use and efficiency of the spray-dryer for each evaluated polymer, we planned a fractional
experiment. Table 1 shows the 16 experiments proposed by the experimental design, the response variables cor-
responding to each experiment, and the global desirability values. The effect of each variable on the orange EO
encapsulation can be verified in Fig. 1.
Based on the experimental data, we found that, when using gelatin as the biopolymer (Fig. 1a,c), the variables
feed flow rate (B), air injection flow (D), wall material content (E), EO/polymer ratio (G), and surfactant (H)
showed negligible effects (less than 10%), Fig. 1c. Only the inlet air temperature (A), drying air-flow (C), and
adjuvant/polymer ratio (F) showed significant effects on the drying process (Fig. 1c).

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 2

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 1.  Graphic representation of the main differences in the investigated independent variables using gelatin
(a) and lignin (b) as encapsulation materials. Percentage of each difference when using gelatin (c) and lignin
(d) as the encapsulation materials. Evaluated variables: A: inlet air temperature; B: feed flow rate; C: drying air-
flow; D: air injection flow; E: wall material content; F: adjuvant/polymer ratio; G: essential oil/polymer ratio; H:
surfactant.

The positive effect of temperature (Fig. 1a) indicates better oil content and higher powder recovery with higher
drying air temperatures. Similar results were described for the encapsulation of essential oils from ­orange14,
­oregano19, and r­ osemary20 when the authors used temperatures higher than 130 °C. The increase in temperature
favored the drying process by reducing the surface tension and the viscosity of the material, assisting in the
droplets ­formation21,22.
In this study, the drying air-flow also showed a positive effect on the oil content and powder recovery, in
which an increase in the flow rate of heated air in the drying chamber favored the atomization process (Fig. 1a).
In 2003, Bruschi et al.23 demonstrated that higher drying air flows allow the pulverized colloidal suspension to
evaporate before contacting the surface of the drying ­chamber24, also improving the formation of products with
lower moisture content.
The adjuvant/polymer ratio was the only independent variable that had a negative effect (Fig. 1a); that is,
the addition of adjuvant in the formulation did not improve the atomization process. In general, the use of
technological adjuvants ("auxiliary dryers"), such as colloidal silicon dioxide, maltodextrins, and cyclodextrins,
influence the powder recovery of the drying process and the redispersion of the powdered products in the col-
loidal ­suspension25–27.
For the products obtained using lignin as the encapsulation material, we verified that the drying air-flow
and the feed flow rate showed the greatest influences on the process, resulting in a better atomization efficiency
(Fig. 1b,d). The combination of the lower sample transfer rate (0.15 L h−1) and the higher drying air flow rate
(536 L h−1) resulted in better powder recovery and oil content in this atomization process (Table 1).
Comparing the results obtained for both biopolymers, we can observe that the highest oil content ranged
between 30 and 90% (w w−1) (Table 1). The oil content of up to 41% was reported using gelatin as wall material
in propolis e­ ncapsulation23. In the literature, we observed values ranging from 71 to 98% in the encapsulation
of orange essential oil by spray-drying process using Arabic gum, or polymeric combinations, r­ epectively14,15.
We could also verify that lignin promoted the best powder recovery (Table 1). We obtained powder recov-
ery between 44 and 72% (w w−1) for gelatin and lignin, respectively. Theses powder recoveries were obtained
in experiments with global disability values higher than 0.70 (Table 1). In your turn, Márquez-Gómez et al.
(2018)14 described powder recoveries of 29, 36, and 73% by using maltodextrin, hydrolyzed protein, and rice
starch, respectively, for encapsulation of orange essential oil. These data indicate that gelatin and lignin are a
viable alternative for the encapsulation of orange EO.
Additionally, we evaluated the encapsulation efficiency (Table 1). Through these data, we observed EO recov-
eries ranging from 5 to 97% (w w−1) according to each experimental condition proposed by experimental design.
Values higher than 90% confirm the retention capacity of orange essential oil in gelatin and lignin microparticles
when they were applied as wall materials. Moreover, these data suggest that there was only a small loss of EO
during emulsion production and atomization drying, indicating the efficiency of the proposed method.

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 3

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 2.  Microphotographs of the microparticles loaded with orange essential oil produced by spray-drying
using gelatin (A) and lignin (B) as the biopolymers (with × 5,000 of magnification).

In this way, for the next steps of this study, emulsions with both biopolymers were prepared without the
addition of a surfactant and colloidal adjuvant. They were subjected to atomization drying with a flow transfer
rate of 0.15 L h−1, drying air-flow rate of 536 L h−1, and inlet air temperature of 150 °C.

Microparticle characterization.  The microparticles obtained by using gelatin presented a microspheri-

cal structure with a smooth and continuous surface, which is fundamental to prevent the penetration of gases
and protect the nucleus (Fig. 2a)11. These morphological characteristics relate to the ability of gelatin to form a
continuous film around the droplets of essential oil before the drying p ­ rocess28,29. The gelatin microparticles also
did not present cracks, which could cause loss of the volatile fractions of the microencapsulated c­ ompounds11
because of their ­permeability20.
When we used lignin as the biopolymer, the microparticles presented an irregular and rough structure char-
acteristic of the biopolymer used (Fig. 2b)11. Although lignin has emulsion-stabilizing ­properties12, it showed
irregular morphological characteristics that can be a challenge for its application in microencapsulation processes
due to the wide diversity of functional groups, which may vary according to its botanical and industrial o­ rigins12.
Although both polymers showed some agglomeration of microparticles, it was still possible to identify indi-
vidual microparticles, indicating that the agglomerates were formed after the complete drying of each micropar-
ticle. Similar results were obtained in the encapsulation of chia and oregano essential o ­ ils19,30.
The microparticles also presented a heterogeneous size distribution showing a peak of frequency under
4.0 μm. Particles containing gelatin as wall material showed a larger average particle size (3.69 μm) compared to
particles containing lignin (2.71 μm) (Supplementary Material Fig. S1). Moreover, more than 50% of the gelatin
particles presented sizes 4 and 10 μm. The largest average particle size may be related to the higher solubility
and emulsification capacity of this polymeric material. Thus, lignin, which has a lower solubility in an aqueous
medium than gelatin, produces smaller particles with matrix c­ haracteristics10.
According to the data that we obtained through our experimental design and the morphological analyses, we
observed that the microparticles obtained by using gelatin and lignin as biopolymers were able to load the orange
essential oil. Although the best results of oil content were obtained with gelatin, we would like to highlight the
lignin formulations, which also showed good results. To the best of our knowledge, this is the first work describ-
ing the microencapsulation of essential oils by using lignins, a waste product from the pulp and paper industries.
Moreover, lignin can act as a natural ­antioxidant31 and has the potential to incorporate antimicrobial activity
and photostability in the final microencapsulated product. Additionally, after the method was optimized, the
powder recovery of the final product was almost three times higher with lignin than that obtained with gelatin.

Quantitative analyses of the microencapsulated essential oil by gas chromatography.  In

addition to the encapsulation stage, the quantification of the encapsulated essential oil is an important step to
evaluate the quality of the final product. The different intrinsic properties between the essential oils (low molecu-
lar weight and have high vapor ­pressure1) and the biopolymers (macromolecules with properties that are the
opposite of essential ­oils32,33) present a challenge in developing a robust quantitative method that is compatible
with both materials. Among the different techniques for sample preparation and analyses, the quantification by
gas chromatography with headspace extraction is a good alternative, due to reduce the possibility of evaporation
losses during sample preparation ­steps34, ensuring the transference only of volatile organic compounds.
Thus, to evaluate the performance of the developed quantitative method by headspace gas chromatography
(headspace GC-FID) and to ensure its reliability, we carried out several analyses that investigated parameters
such as selectivity, matrix effects, limits of detection and quantification, linearity, precision, and accuracy.

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 4

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 3.  Chromatograms of the microparticles of gelatin (A) and lignin (C) loaded with d-limonene
(1.0 µg mL−1), and microparticle controls of gelatin (B) and lignin (D) without d-limonene. Identification of the
peaks: (1) d-limonene and (2) menthol (IS).

Selectivity.  The chromatograms of the control samples (orange essential oil-free microparticles) and the gel-
atin and lignin microparticles containing orange essential oil showed no interference peaks from the matrices or
the internal standard that affected the retention time of the analyte (Fig. 3). Therefore, the developed method was
shown to be selective for the analysis of d-limonene and could distinguish the response of the analyte of interest
(d-limonene) from the other responses produced by the components from the matrices (i.e., the biopolymers)35.

Evaluation of matrix effects in the d‑limonene analysis by headspace gas chromatogra‑

phy.  In this study, we investigated the matrix effects that are directly related to the headspace extraction step.
To evaluate the matrix effects on the headspace extraction of d-limonene, we carried out a matrix-matched
calibration. The influence of co-extractives on the d-limonene headspace extraction could be observed through
the slope differences among the analytical calibration curves, which were obtained for lignin, gelatin, and their
combination with a colloidal adjuvant, as illustrated in Fig. 4.
The ratios between the angular coefficients obtained in the calibration curves in the matrix and solvent (Sup-
plementary Table S2) were higher than 1.00, indicating a positive matrix effect. The matrix effects could also be
observed by variations in the quantification of the analyte in the powdered products according to the prepared
calibration curve in the matrix or pure solvent (Supplementary Fig. S2). When we did not include the matrix
effects, the essential oil recoveries were greater than 150%, for example, in the gelatin formulations. Clearly, this
is caused by matrix effects. Thus, when we included the interference of the matrix components, there was vari-
ation in the chromatographic signal for d-limonene, with variations as high as 96% and 89% for the gelatin and
lignin formulations, respectively.
These data indicate that the presence of the matrix components promotes greater extraction in the headspace
with an increased chromatographic signal for d-limonene. This finding may be explained by partial solvent
vaporization during headspace extraction. In the presence of the matrix components, the solvent may interact
with and/or dissolve other molecules present on the matrices, decreasing its concentration in the vapor phase,
and assisting in the release and extraction of d-limonene (thereby increasing the coefficient of volatilization of
d-limonene).
Therefore, the quantitative analyses of microparticles loaded with d-limonene were carried out using analyti-
cal calibration curves prepared in the presence of the matrices to minimize their effects and to guarantee the
accuracy of the analytical method.

Linearity and limits of detection and quantification of the GC‑FID analytical method.  The lin-
earity of the proposed method was evaluated according to the coefficient of determination (r2) from the calibra-
tion curves prepared in the presence of the biopolymer matrices of gelatin and lignin, and in the acetone solvent
(Fig. 4). The calibration curves were linear, and covered a range from 0.025 to 3.00 µg mL−1 with coefficients of
determination (r2) greater than 0.99 regardless of the matrix effects, indicating a good correlation between the
area of the d-limonene peak and its ­concentration36.
In this study, we also found that when using gelatin as the matrix, the calibration curve presented a higher ana-
lytical sensitivity to detect variations in the concentration of d-limonene. The analysis of variance (ANOVA), the
graphs of the residues for the calibration curve samples in relation to the adjusted regression and the Cochran’s
C test for homogeneity (95% confidence level) are presented in the supplementary material (Supplementary
Table S3 and S4 and Supplementary Fig. S3).
The LOD and LOQ values obtained (Table 2) verified that the analytical calibration curve prepared by using
extracts from the gelatin matrix showed values of LOD and LOQ that were approximately twice those of the
values obtained in the solvent. These data confirmed the stronger matrix effect on the analytical sensitivity when

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 5

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 4.  Analytical curves for d-limonene prepared in solvent (acetone) and matrices. (a) Gelatin and
gelatin:Aerosil. (b) Lignin and lignin:Aerosil.

%R ± RSD
LOD LOQ Concentration (µg mL−1)
Matrix Equation r2 (mg g−1) 0.30a 1.2a 2.4b
Acetone y = 0.941x − 0.009 0.999 0.051 0.153 – – –
Gelatin y = 1.972x − 0.086 0.998 0.024 0.073 103 ± 3.3 105 ± 5.5 111 ± 7.1
Lignin y = 1.212x − 0.025 0.995 0.039 0.119 99.1 ± 3.1 109 ± 6.0 103 ± 13

Table 2.  Limit of detection (LOD) and quantification (LOQ), the linearity of the method, and power
recovery percentages of d-limonene after extraction of gelatin and lignin samples spiked with three different
concentrations. a Three repetitions; bseven repetitions. r2 coefficient of determination, LOD limit of detection,
LOQ limit of quantification, R recovery, RSD relative standard deviation.

­ iopolymer37. The sensitivity and linearity of the method were also demonstrated by the low
gelatin was used as b
values of the obtained LOD, which were significantly lower than the minimum necessary for the quality control
of the microparticles loaded with d-limonene.

Precision and accuracy.  The ability of the developed analytical method to present the concentration–
response relationship (linearity) and reproducibility was evaluated by systematically repeating analyses of
known samples.
The precision was expressed by the Relative Standard Deviation (RSD) obtained for seven identical headspace
extractions of fortified samples with 2.4 µg mL−1 of d-limonene, using lignin and gelatin as the biopolymers.
Table 2 describes the RSD values, which were lower than 15% for both biopolymers (values of 7.1 and 13% for
gelatin and lignin, respectively). These results demonstrate the reproducibility of the method. RSD values that
are close to or smaller than the recommended (RSD < 20%) indicate the low influence of random errors, which
­ ethod38,39.
is considered to be acceptable in ensuring the precision of the m

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 6

Vol:.(1234567890)
www.nature.com/scientificreports/

The accuracy of the analytical method was evaluated by comparing the nominal and experimental values
that were obtained by the quantification d-limonene in the samples within the linear range of the calibration
curves, at three different concentration levels (low, medium, and high), which were prepared independently
in triplicate. Table 2 shows the accuracy and RSD values of the three concentration levels. For all matrices, the
method showed good performance, with accuracy between 99 and 111%, which indicated an agreement among
the experimental and theoretical values and a low influence of systematic errors.

Conclusion
In this work, formulations to encapsulate orange essential oil were successfully obtained using spray-drying
process. Fractional factorial design indicated that the main variables that should be controlled to obtain the best
powder recovery and oil content were the sample flow transfer rate, drying air-flow, and temperature.
Working under the optimized conditions, concluded that it is possible to prepare microparticles by spray-
drying using gelatin and lignin as biopolymers. It was possible to prepare microparticles showing microspherical
morphology, with the average particle size of less than 4 µm (gelatin) and 3 µm (lignin), with an oil content of
90% (w w−1) for gelatin and powder recovery of 72% (w w−1) for lignin as wall material.
The quantification method development and validation were compatible with the intrinsic features of the
essential oil and the biopolymers. The matrix effects were addressed by using matrix-matched calibration curves.
In doing so, the analytical method was validated, showing linearity, selectivity, precision, and accuracy for the
quantitative analysis of d-limonene. Additionally, the analytical method presented advantages such as its opera-
tional simplicity. The analytical method also assisted in the optimization of the experimental variables for the
production of microparticles loaded with essential oil by providing quantitative data.

Experimental section
Chemicals.  Stock standard solutions of d-limonene (99% w w−1), purchased from Sigma-Aldrich (St. Louis,
United States of America), and menthol as an internal standard (Arora Chemicals, Sao Marcos, Brazil) were
prepared in acetone at 5.0 and 8.0 mg mL−1, respectively. Both solutions were stored at − 4 °C.
The solvents used were acetone of HPLC grade (Panreac, Barcelona, Spain) and ultrapure water (Milli-Q,
Millipore). Gelatin type B (Synth, Sao Paulo, Brazil) and lignin (provided by Suzano S/A, Sao Paulo, Brazil) were
used as the encapsulating materials. The surfactant sorbitan monostearate (Span 60, Sigma-Aldrich, St. Louis,
United States of America) and the adjuvant colloidal silicon dioxide (Aerosil 200, Evonik, Essen, Germany) were
also used in the formulations.

Sample.  The orange essential oil, produced by cold pressing the orange peel, was obtained from the region
of Santa Cruz do Rio Pardo—SP (Agroterenas S/A Citrus, Brazil). This EO was used as the volatile core material.
The specifications of the EO were as follows: density at 25 °C: 0.842 g mL−1, the refractive index at 20 °C: 1.473,
and aldehydes: 1.4%.
To evaluate the volatile composition, the volatile compounds were analyzed by a gas chromatograph hyphen-
ated with a sequential mass spectrometer (Shimadzu Corporation, Japan TQ8040). In these analyses, we used
a capillary column (Rtx-5MS, Restek) with a stationary phase consisting of 5% diphenyl and 95% dimethyl-
polysiloxane (30 m × 0.25 mm d.i. × 0.25 µm film thickness) and Helium 5.0 as carrier gas (1 mL min−1). The
chromatographic conditions were as follows: the injector temperature was 220 °C; the initial oven temperature
for the column was 60 °C (1 min), then it was heated at a rate of 3 °C min−1 to 230 °C and maintained at this
temperature for 1 min. The interface transfer temperature was 240 °C; the injection mode was split (100:1); the
injected sample volume was 1.0 µL. The electron impact mass spectra were recorded at 70 eV ionization energy,
with the acquisition quadrupole mass ranging from 40 to 700 a.m.u., in scanning mode, with 0.3 scans per sec-
ond. The volatile organic compounds were identified by comparison with a library of spectral data (NIST 17.0),
the literature data described by Adams (2009)40, and the retention index as proposed by Van den Dool e K ­ ratz40.

Orange essential oil microencapsulation.  Multivariate optimization was applied to determine the
optimal condition for drying and encapsulating orange EO by atomization (Spray-drying). For this purpose, we
employed fractional factorial design with eight independent variables ­(28–4), totaling 16 experiments, with the
experiments randomly conducted to minimize the effects of unexpected variability in the observed responses
due to systematic errors. The following variables were selected to optimize the spray-dryer performance, with
high and low increments: inlet air temperature, feed flow rate, drying air-flow, air injection flow (Aspirator), wall
material content, adjuvant/polymer ratio, essential oil/polymer ratio, and surfactant. The levels evaluated are
described in Table 3. Those parameters were selected from previous studies in our research group, and by seeks
in ­literature14,19,41,42. We evaluated the use of gelatin and lignin as the biopolymers.
In this way, colloidal suspensions were progressively produced by the addition of known quantities of biopoly-
mer (mg), surfactant (mg), colloidal adjuvant (mg), and essential oil (mg) into 30 mL of ultrapure water in a
125 mL Erlenmeyer flask (Table 1). The system was maintained under magnetic stirring at 20,000 rpm for 60 s
using a disperser (Ultra-Turrax IKA T10 basic, Wilmington, United States America) at 20 °C (± 1 °C).
The colloidal suspensions were immediately transferred to a spray-dryer (Mini Spray Dryer BÜCHI, B 290,
Flawil, Switzerland) equipped with a drying chamber (500 mm × 200 mm) and a nozzle atomizer (0.7 mm). The
resulting dried products were collected, kept glass desiccator during 48 h, and then stored under refrigeration
(8 °C).
The powder recovery (% w w−1) in the encapsulation process was calculated according to the Eq. (1):

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 7

Vol.:(0123456789)
 www.nature.com/scientificreports/

Levels
Variables − 1 + 1
A Inlet air temperature (°C) 110 150
B Feed flow rate (L h−1) 0.15 0.45
C Drying air flow (L h−1) 301 536
D Air injection flow ­(m3 h−1) 8 35
E Wall material content (% w v−1) 5 10
F Adjuvant/polymer ratio (w w−1), 1:0 1:1
G Essential oil/polymer ratio (w w−1) 1:1.78 1:3.56
H Surfactant (mg) 0 200
Coded levels
Experiment A B C D E F G H
1 − 1 − 1 − 1 − 1 − 1 − 1 − 1 − 1
2 + 1 − 1 − 1 − 1 + 1 + 1 + 1 − 1
3 − 1 + 1 − 1 − 1 + 1 + 1 − 1 + 1
4 + 1 + 1 − 1 − 1 − 1 − 1 + 1 + 1
5 − 1 − 1 + 1 − 1 + 1 − 1 + 1 + 1
6 + 1 − 1 + 1 − 1 − 1 + 1 − 1 + 1
7 − 1 + 1 + 1 − 1 − 1 + 1 + 1 − 1
8 + 1 + 1 + 1 − 1 + 1 − 1 − 1 − 1
9 − 1 − 1 − 1 + 1 − 1 + 1 + 1 + 1
10 + 1 − 1 − 1 + 1 + 1 − 1 − 1 + 1
11 − 1 + 1 − 1 + 1 + 1 − 1 + 1 − 1
12 + 1 + 1 − 1 + 1 − 1 + 1 − 1 − 1
13 − 1 − 1 + 1 + 1 + 1 + 1 − 1 − 1
14 + 1 − 1 + 1 + 1 − 1 − 1 + 1 − 1
15 − 1 + 1 + 1 + 1 − 1 − 1 − 1 + 1
16 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1

Table 3.  Coded levels of the independent variables, and matrix representation of the fractional factorial
design ­(28–4). (− 1) lower level; (+ 1) higher level.

WS
Powder Recovery (%) = × 100 (1)
WS0
where ­WS is the mass of dried material collected after drying in the spray-dryer and W
­ S0 is the initial quantity
of solids in the sprayed emulsion volume.

Oil content and encapsulation efficiency.  The total orange essential oil quantity in the microencapsu-
lated products that were obtained by spray-drying was evaluated by quantifying the d-limonene content present
in the internal phase (the encapsulated nucleus) and adsorbed onto the surface of the particles after solubiliz-
ing 25.0 mg of the dried material in 1,000 µL of acetone for 30 min. To 800 μL of the resulting mixture, 100 μL
of menthol (1,000 µg mL−1, IS) was added. The samples were extracted by static headspace (SH). SH analysis
was performed using a PALSyr HS 2.5 mL for combi-PAL. To do so, 10 µL of the sample was transferred into a
headspace vial (10 mL) and homogenized for 15 min at 75 °C and 500 rpm. Subsequently, 1,000 µL of the vapor
phase was injected into GC.
Quantitative GC analyses were carried out using a Shimadzu (GC 2010 Plus) apparatus coupled with a Flame
Ionization Detector (FID). Analyses were carried out using a ZB-Wax capillary column (30 m × 0.25 mm i.d.;
0.25 µm film thickness) coated with polyethylene glycol (Phenomenex). The programmed oven temperature
was 40 °C for 1 min, raised at 5 °C min−1 to 170 °C and held for 1 min; the injector temperature was 170 °C; the
carrier gas was Helium 5.0 at a constant flow rate of 1 mL min−1; the injection mode was split (15:1). Synthetic
air (300 mL min−1), hydrogen (40 mL min−1) and nitrogen (30 mL min−1) were used for FID.
To evaluate the content of the essential oil in the powder products, at the end drying process (oil content
%, w w−1), we determined the concentration of d-limonene using the data from an analytical curve obtained
by the analyses of standard solutions in concentrations ranging from 0.025 to 3.00 µg mL−1. The encapsulation
efficiency (% w w−1) was obtained comparing the oil content, in the powder products to the used quantity of OE
in the preparation of each emulsion. The oil content (% w w−1) and the encapsulation efficiency (% w w−1) were
calculated using the Eqs. (2) and (3), respectively:
WE
Oil Content (%) = × 100 (2)
WS

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 8

Vol:.(1234567890)
www.nature.com/scientificreports/

WE
Encapsulation Efficiency (%) = × 100 (3)
WE0
where ­WE is the mass of EO obtained after drying and W
­ E0 is the mass of the EO used during the preparation
of emulsions.

Global desirability function for the development of essential oil‑loaded formulations.  To

relate the levels of the dependent variables (powder recovery and oil content) and find the best conditions that
produced the most desirable responses, we used the global desirability ­function25. This function allows for the
normalization of the values obtained from the individual desirability values ­(di)25,43,44 and obtains the global
desirability value (D) for each experiment by calculating the geometric average of the ­di ­values44. The desirability
values are on a scale from 0 to 1, where values closer to 1 are considered ­desirable43. The desirability values were
also used to calculate the main effects and interaction of the variables for the microencapsulation of orange EO.

Microparticle morphology evaluation.  The surface morphology of the microparticles was visualized
using scanning electronic microscopy (SEM). The powdered samples were fixed to a double-sided adhesive car-
bon tape mounted on SEM stubs with a diameter of 12 mm (Koch, Sao Paulo, Brazil), and then coated with a thin
layer of gold/palladium in a vacuum. The images were magnified to 200 to 50,000 × under an FEI Inspect S50
microscope operating at 25 kV41. The particle size was measured by analyzing the images with ImageJ ­software45.
The particle size distribution was determined by measuring the diameter of approximately 300 particles. The
resulting values were plotted on a histogram and adjusted with a Gaussian function.

Analytical method validation.  The selectivity, linearity, limits of detection (LOD) and quantification
(LOQ), accuracy, and precision are parameters that were used to evaluate the analytical method for the quanti-
fication of the orange essential ­oil38,39,46.
The linearity was verified by preparing standard solutions of increasing concentrations of the analyte
d-limonene in solution (0.025, 0.075, 0.250, 0.750, 2.00 and 3.00 µg mL−1) and also using a solid matrix (gel-
atin, gelatin and aerosil, lignin, and lignin and aerosil) as the dispersal medium, and an internal standard
(1.0 µg mL−1). Thus, for each condition, 10 µL of the standard solution was transferred into a headspace vial
(10 mL) and homogenized for 15 min at 75 °C and 500 rpm. Subsequently, 1,000 µL of the vapor phase was
injected into the GC.
After the analysis of these solutions, a graph was constructed relating the ratio between the area of the analyte
and the internal standard to their respective concentration used in the calibration. The linearity was evaluated by
the determination coefficient (r2) calculated by linear regression and verified by analysis of variance regarding
regression, accuracy, and analytical recovery. The evolution of the matrix co-extractives in the chromatographic
signal was evaluated by the relationship between the area of the analyte in the pure solvent (acetone) and the
area obtained using the standard prepared in aqueous extracts of the matrix (matrix-matching), where the
matrix effect (%) = (Āmatrix  – Āsolvent/Āsolvent) × 100). Extracts of the matrix (gelatin, gelatin and Aerosil, lignin,
and lignin and Aerosil) were also used as a dispersal medium for the preparation of the curves to determine the
matrix effects.
The selectivity was observed by evaluating the chromatograms of the matrix that were obtained after the
extraction of the components in the analyte-free matrix compared to the chromatograms of the extracts of the
microparticles that were prepared and analyzed according to the optimized procedure.
The LOD and LOQ for d-limonene were determined for each matrix by successively diluting the stock solu-
tion of d-limonene and by the parameters of the analytical curve, using the following equations: LOD = (3.3δ)/S
and LOQ = (10δ)/S, where δ is the standard deviation of the calculated controls by analyzing the noise of three
samples and S is the slope of the calibration curve.
The method precision was determined by calculating the relative standard deviation and the coefficient of
variation of seven formulations in identical extractions of the orange EO at a concentration of 2.4 µg mL−1. The
extraction recoveries (%R) were determined in samples of microparticles containing d-limonene at different
levels: 0.30, 1.2, and 2.4 µg mL−1. The accuracy was evaluated by the recovery of the analyte.

Statistical analysis.  Analysis of variance (ANOVA), multiple comparisons of means by the Tukey test (5%
probability), and the Cochran test (5% probability) were performed using SPSS software (Statistical Package for
Social Science for Windows), version 17.0.0 (2008).

Received: 10 February 2020; Accepted: 24 June 2020

References
1. Pavela, R. & Benelli, G. Essential oils as ecofriendly biopesticides? challenges and constraints. Trends Plant Sci. 21, 1000–1007
(2016).
2. Isman, M. B., Miresmailli, S. & MacHial, C. Commercial opportunities for pesticides based on plant essential oils in agriculture,
industry and consumer products. Phytochem. Rev. 10, 197–204 (2011).
3. Regnault-Roger, C., Vincent, C. & Arnason, J. T. Essential oils in insect control: low-risk products in a high-stakes world. Annu.
Rev. Entomol. 57, 405–424 (2012).

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 9

Vol.:(0123456789)
 www.nature.com/scientificreports/

4. Ferhat, M. A., Meklati, B. Y., Smadja, J. & Chemat, F. An improved microwave Clevenger apparatus for distillation of essential oils
from orange peel. J. Chromatogr. A 1112, 121–126 (2006).
5. Rezzadori, K., Benedetti, S. & Amante, E. R. Proposals for the residues recovery: orange waste as raw material for new products.
Food Bioprod. Process. 90, 606–614 (2012).
6. Rodríguez, A. et al. Impact of d-limonene synthase up- or down-regulation on sweet orange fruit and juice odor perception. Food
Chem. 217, 139–150 (2017).
7. Asbahani, A. E. et al. Essential oils: from extraction to encapsulation. Int. J. Pharm. 483, 220–243 (2015).
8. Knowles, A. Recent developments of safer formulations of agrochemicals. Environmentalist 28, 35–44 (2008).
9. Venkatesan, P., Manavalan, R. & Valliappan, K. Microencapsulateation: a vital technique in novel drug delivery system. J. Pharm.
Sci. Res. 1, 26–35 (2009).
10. Madene, A., Jacquot, M., Scher, J. & Desobry, S. Flavour encapsulation and controlled release—a review. Int. J. Food Sci. Technol.
41, 1–21 (2006).
11. Aghbashlo, M., Mobli, H., Rafiee, S. & Madadlou, A. Optimization of emulsification procedure for mutual maximizing the encap-
sulation and exergy efficiencies of fish oil microencapsulation. Powder Technol. 225, 107–117 (2012).
12. Witzler, M. et al. Lignin-derived biomaterials for drug release and tissue engineering. Molecules 23, 1–22 (2018).
13. Sutaphanit, P. & Chitprasert, P. Optimisation of microencapsulation of holy basil essential oil in gelatin by response surface meth-
odology. Food Chem. 150, 313–320 (2014).
14. Márquez-Gómez, M., Galicia-García, T., Márquez-Meléndez, R., Ruiz-Gutiérrez, M. & Quintero-Ramos, A. Spray-dried micro-
encapsulation of orange essential oil using modified rice starch as wall material. J. Food Process. Preserv. 42, e13428 (2018).
15. de Souza, H. J. B. et al. Utility of blended polymeric formulations containing cellulose nanofibrils for encapsulation and controlled
release of sweet orange essential oil. Food Bioprocess Technol. 11, 1188–1198 (2018).
16. Carmona, P. A. O., Tonon, R. V., da Cunha, R. L. & Hubinger, M. D. Influence of emulsion properties on the microencapsulation
of orange essential oil by spray drying. J. Colloid Sci. Biotechnol. 2, 130–139 (2013).
17. Mira, B., Blasco, M., Berna, A. & Subirats, S. Supercritical ­CO2 extraction of essential oil from orange peel. Effect of operation
conditions on the extract composition1. J. Supercrit. Fluids 14, 95–104 (1999).
18. Farhat, A. et al. Microwave steam diffusion for extraction of essential oil from orange peel: kinetic data, extract’s global yield and
mechanism. Food Chem. 125, 255–261 (2011).
19. Alvarenga Botrel, D. et al. Evaluation of spray drying conditions on properties of microencapsulated oregano essential oil. Int. J.
Food Sci. Technol. 47, 2289–2296 (2012).
20. Fernandes, R. V. D. B., Borges, S. V. & Botrel, D. A. Gum arabic/starch/maltodextrin/inulin as wall materials on the microencap-
sulation of rosemary essential oil. Carbohydr. Polym. 101, 524–532 (2014).
21. Alves, S. F. et al. Microencapsulation of essential oil from fruits of pterodon emarginatus using gum arabic and maltodextrin as
wall materials: composition and stability. Dry. Technol. 32, 96–105 (2014).
22. Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A. & Saurel, R. Applications of spray-drying in microencapsulation of food
ingredients: an overview. Food Res. Int. 40, 1107–1121 (2007).
23. Bruschi, M. L., Cardoso, M. L. C., Lucchesi, M. B. & Gremião, M. P. D. Gelatin microparticles containing propolis obtained by
spray-drying technique: preparation and characterization. Int. J. Pharm. 264, 45–55 (2003).
24. Oliveira, O. W. & Petrovick, P. R. Secagem por aspersão (spray drying) de extratos vegetais: bases e aplicações. Braz. J. Pharmacogn.
20, 641–650 (2010).
25. Ferreira, S. L. C. et al. Box–Behnken design: an alternative for the optimization of analytical methods. Anal. Chim. Acta 597,
179–186 (2007).
26. Tewa-Tagne, P. et al. Spray-drying nanocapsules in presence of colloidal silica as drying auxiliary agent: formulation and process
variables optimization using experimental designs. Pharm. Res. 24, 650–661 (2007).
27. Da Costa, J. M. G. et al. Matrix structure selection in the microparticles of essential oil oregano produced by spray dryer. J. Micro-
encapsul. 30, 717–727 (2013).
28. Moretti, M. D. L., Sanna-Passino, G., Demontis, S. & Bazzoni, E. Essential oil formulations useful as a new tool for insect pest
control. AAPS PharmSciTech 3, E13 (2002).
29. Yoshii, H. et al. Flavor release from spray-dried maltodextrin/gum arabic or soy matrices as a function of storage relative humidity.
Innov. Food Sci. Emerg. Technol. 2, 55–61 (2001).
30. Rodea-González, D. A. et al. Spray-dried encapsulation of chia essential oil (Salvia hispanica L) in whey protein concentrate-
polysaccharide matrices. J. Food Eng. 111, 102–109 (2012).
31. Sipponen, M. H., Lange, H., Crestini, C., Henn, A. & Österberg, M. Lignin for nano- and microscaled carrier systems: applications,
trends, and challenges. Chemsuschem 12, 2038 (2019).
32. Vieira, M. G. A., Da Silva, M. A., Dos Santos, L. O. & Beppu, M. M. Natural-based plasticizers and biopolymer films: a review. Eur.
Polym. J. 47, 254–263 (2011).
33. Yates, M. R. & Barlow, C. Y. Life cycle assessments of biodegradable, commercial biopolymers—a critical review. Resour. Conserv.
Recycl. 78, 54–66 (2013).
34. Cavalli, J. F., Fernandez, X., Lizzani-Cuvelier, L. & Loiseau, A. M. Comparison of static headspace, headspace solid phase micro-
extraction, headspace sorptive extraction, and direct thermal desorption techniques on chemical composition of french olive oils.
J. Agric. Food Chem. 51, 7709–7716 (2003).
35. Taverniers, I., De Loose, M. & Van Bockstaele, E. Trends in quality in the analytical laboratory. II. Analytical method validation
and quality assurance. TrAC Trends Anal. Chem. 23, 535–552 (2004).
36. Shabir, G. A., John Lough, W., Arain, S. A. & Bradshaw, T. K. Evaluation and application of best practice in analytical method
validation. J. Liq. Chromatogr. Relat. Technol. 30, 311–333 (2007).
37. Gustavo González, A. & Ángeles Herrador, M. A practical guide to analytical method validation, including measurement uncer-
tainty and accuracy profiles. TrAC Trends Anal. Chem. 26, 227–238 (2007).
38. Food and Drug Administration. Guidance for Industry: Bioanalytical Method Validation. https​://www.fda.gov/regul​atory​-infor​
matio​n/searc​h-fda-guida​nce-docum​ents (2018).
39. Agência Nacional de Vigilância Sanitária. RDC N ­ o 166, de 24 de julho de 2017. https​://porta​l.anvis​a.gov.br/legis​lacao​ (2017).
40. Adams, R. Identification of essential oil components by gas chromatography/quadrupole mass spectroscopy 54–788 (2009)
41. de Fernandes, R. V. B. et al. Microencapsulation of rosemary essential oil: characterization of particles. Dry. Technol. 31, 1245–1254
(2013).
42. Roccia, P., Martínez, M. L., Llabot, J. M. & Ribotta, P. D. Influence of spray-drying operating conditions on sunflower oil powder
qualities. Powder Technol. 254, 307–313 (2014).
43. Lazic, Z. I. R. Design of Experiments in Chemical Engineering A Practical Guide (Wiley, Weinhem, 2004). https​://doi.
org/10.1002/35276​04162​.
44. Ibrahim Silva, P., Stringheta, P. C., Teof́ilo, R. F. & De Oliveira, I. R. N. Parameter optimization for spray-drying microencapsula-
tion of jaboticaba (Myrciaria jaboticaba) peel extracts using simultaneous analysis of responses. J. Food Eng. 117, 538–544 (2013).
45. Abràmoff, M. D., Hospitals, I., Magalhães, P. J. & Abràmoff, M. Image processing with ImageJ. Biophotonics Int. 11, 36–42 (2007).
46. Thompson, M., Ellison, S. L. R. & Wood, R. Harmonized guidelines for single-laboratory validation of methods of analysis (IUPAC
Technical Report). Pure Appl. Chem. 74, 835–855 (2002).

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 10

Vol:.(1234567890)
www.nature.com/scientificreports/

Acknowledgements
The authors are grateful for the financial support received from the National Council for Scientific and Techno-
logical Development—CNPq (403302/2013-7; 306109/2018-2; 429404/2018-2), the São Paulo Research Foun-
dation—FAPESP (2012/25266-9; 2018/21201-8; 2014/50918-7), Coordination for the Improvement of Higher
Education Personnel—CAPES (Fomento 001), and by the infrastructure and support of the Department of
Chemistry of the Federal University of São Carlos. We are also grateful to Suzano S/A, and Agroterenas S/A
Citrus for the provided lignin and orange essential oil, respectively.

Author contributions
M.R.F. conceived the study, analyzed and interpreted the data, and critically read and revised the manuscript.
M.C.S.A. designed the experiments, acquired, analyzed, and interpreted the data, and drafted the manuscript.
M.F.G.F.S. and J.B.F. participated in the data discussion. All of the authors contributed to data analysis and
reviewed and approved the final manuscript.

Competing interests 
The authors declare no competing interests.

Additional information
Supplementary information is available for this paper at https​://doi.org/10.1038/s4159​8-020-68823​-4.
Correspondence and requests for materials should be addressed to M.R.F.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access  This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.

© The Author(s) 2020

Scientific Reports | (2020) 10:11799 | https://doi.org/10.1038/s41598-020-68823-4 11

Vol.:(0123456789)